The Cellular Chaperone Heat Shock Protein 90 ... - Journal of Virology

3 downloads 20660 Views 307KB Size Report
Dec 15, 2004 - tion we use Flock House virus (FHV), the best-studied mem- ber of the .... responsive luciferase expression plasmid pS2GRE-LUC were generously pro- vided by Jorge .... used pMT-V5/His/LacZ as a template, the sense primer 5-TAATACGACTCA ...... Sheffield, W. P., G. C. Shore, and S. K. Randall. 1990.
JOURNAL OF VIROLOGY, June 2005, p. 6827–6837 0022-538X/05/$08.00⫹0 doi:10.1128/JVI.79.11.6827–6837.2005 Copyright © 2005, American Society for Microbiology. All Rights Reserved.

Vol. 79, No. 11

The Cellular Chaperone Heat Shock Protein 90 Facilitates Flock House Virus RNA Replication in Drosophila Cells Kathryn M. Kampmueller1 and David J. Miller1,2* Departments of Internal Medicine1 and Microbiology & Immunology,2 University of Michigan Medical School, Ann Arbor, Michigan 48109 Received 15 December 2004/Accepted 19 January 2005

The assembly of viral RNA replication complexes on intracellular membranes represents a critical step in the life cycle of positive-strand RNA viruses. We investigated the role of the cellular chaperone heat shock protein 90 (Hsp90) in viral RNA replication complex assembly and function using Flock House virus (FHV), an alphanodavirus whose RNA-dependent RNA polymerase, protein A, is essential for viral RNA replication complex assembly on mitochondrial outer membranes. The Hsp90 chaperone complex transports cellular mitochondrial proteins to the outer mitochondrial membrane import receptors, and thus we hypothesized that Hsp90 may also facilitate FHV RNA replication complex assembly or function. Treatment of FHV-infected Drosophila S2 cells with the Hsp90-specific inhibitor geldanamycin or radicicol potently suppressed the production of infectious virions and the accumulation of protein A and genomic, subgenomic, and template viral RNA. In contrast, geldanamycin did not inhibit the activity of preformed FHV RNA replication complexes. Hsp90 inhibitors also suppressed viral RNA and protein A accumulation in S2 cells expressing an FHV RNA replicon. Furthermore, Hsp90 inhibition with either geldanamycin or RNAi-mediated chaperone downregulation suppressed protein A accumulation in the absence of viral RNA replication. These results identify Hsp90 as a host factor involved in FHV RNA replication and suggest that FHV uses established cellular chaperone pathways to assemble its RNA replication complexes on intracellular membranes. with the chaperone heat shock protein 70 (Hsp70) and members of the Hsp40 cochaperone family during initial protein folding (6). A select group of cellular proteins also use Hsp90 and its associated cochaperones to complete their maturation. Hsp70 and Hsp90 can function together as part of a larger multichaperone complex, which is connected functionally and physically by various cochaperones (38). The Hsp90 chaperone complex participates in several cellular processes, including vesicle secretion and recycling, protein complex assembly and disassembly, and protein transport. In particular, Hsp90 facilitates mitochondrial preprotein delivery to the outer membrane import receptors in higher eukaryotes (57). To study viral RNA replication complex assembly and function we use Flock House virus (FHV), the best-studied member of the Nodaviridae family (4). FHV is used as a model pathogen to investigate viral capsid formation and genome packaging (44), viral RNA replication and subgenomic synthesis (24, 27, 39), virus-mediated RNA interference (RNAi) suppression (26), and viral RNA replication complex assembly and function (30–32, 56), in part due to its robust replication in multiple hosts, including Saccharomyces cerevisiae (27, 30, 32, 39, 40) and Drosophila melanogaster (31) cells. FHV contains one of the smallest known genomes of any animal RNA virus (4). The 4.5-kb genome is bipartite, with two capped but nonpolyadenylated RNA segments copackaged into a 29-nm nonenveloped icosahedral capsid (47). The larger 3.1-kb RNA species (RNA1) encodes protein A, the FHV RNA-dependent RNA polymerase (RdRp) (4). Protein A is both necessary and sufficient for the assembly of functional viral RNA replication complexes (3, 24, 27, 30, 39). The smaller 1.4-kb RNA species (RNA2) encodes the structural capsid protein, which is essential for virion formation but dispensable for RNA replication

Studies with numerous positive-strand RNA viruses from different families have demonstrated that viral RNA replication occurs within intracellular membrane-associated macromolecular complexes (1), suggesting that the host-pathogen interactions that facilitate the formation of these complexes represent crucial determinants of viral pathogenesis. However, the mechanisms whereby viruses direct their RNA replication complex proteins to the appropriate intracellular membrane, and the cellular components involved in this process, are not well understood. An important step in the assembly of viral RNA replication complexes is the intracellular transport of viral proteins and RNA to the appropriate host membrane compartment. Viral RNA replication complex proteins often contain hydrophobic domains that are essential for their membrane association (30, 43, 52), and thus these viral proteins are susceptible to the same potential folding and aggregation difficulties encountered by cellular proteins with hydrophobic domains (18). Cellular chaperones that facilitate protein folding and transport have been implicated in the replication of many viruses with diverse genome structures and replication strategies (49). While some viruses in the Polyomaviridae and Closteroviridae families encode their own viral-specific chaperone proteins, most viruses use the cellular chaperone machinery to complete their replication (49). The appropriate folding of cellular proteins is essential for their function and stability and is mediated by a group of ubiquitous cytosolic proteins referred to as molecular chaperones (18). Many newly synthesized proteins transiently interact * Corresponding author. Mailing address: Department of Internal Medicine, Division of Infectious Diseases, 5220E MSRB III, 1150 W. Medical Center Drive, Ann Arbor, MI 48109-0640. Phone: (734) 7630565. Fax: (734) 764-0101. E-mail: [email protected]. 6827

6828

KAMPMUELLER AND MILLER

(4). During viral RNA replication, FHV produces a subgenomic 0.4-kb RNA species (RNA3) that is colinear with the 3⬘ end of RNA1. RNA3 encodes protein B, which functions as an RNAi suppressor (26). FHV RNA replication complexes are located on mitochondrial outer membranes in conjunction with 50- to 70-nm membrane-bound spherules in the intermembrane space (31) and are targeted and anchored to mitochondrial outer membranes by protein A via an amino-proximal transmembrane domain and adjacent residues (30, 32). The protein A mitochondrial membrane targeting signal resembles targeting domains present in cellular mitochondrial proteins (30), and thus FHV may use established cellular mechanisms to assemble its viral RNA replication complexes. In this report, we describe the use of pharmacologic and genetic approaches to examine the role of the molecular chaperone Hsp90 in FHV RNA replication in Drosophila S2 cells. We demonstrate that Hsp90 was important for the production of infectious virions and the accumulation of viral RNA and protein A but not for the activity of preformed FHV RNA replication complexes. MATERIALS AND METHODS Cells, virus, and infection protocol. We used Drosophila S2 cells grown at 25°C in Schneider’s insect medium supplemented with 10% heat-inactivated fetal bovine serum, 10 units penicillin per ml, and 10 ␮g streptomycin per ml, unless otherwise indicated. S2 cells were routinely passed every 3 to 4 days at a 1:5 dilution to maintain high density and vigorous cell proliferation. S2 cells were infected with sucrose-gradient purified FHV at a multiplicity of infection (MOI) of 10. Initial studies indicated that FHV-infected S2 cells recapitulated key biochemical, cellular, and morphological features of FHV RNA replication complex assembly and function previously demonstrated with Drosophila DL-1 cells (31), including the mitochondrial localization and temporal appearance of functional RNA replication complexes by 4 h postinfection and the presence of 50- to 70-nm membrane-bound spherules connected to the mitochondrial outer membrane (data not shown). FHV infection does not cause cytolysis in S2 cells (47), and thus we quantitated infectious virions using an immunofluorescence-based assay. Confluent S2 cell monolayers on polyethyleneimine-coated chamber slides were infected with serial virus dilutions, fixed with 4% paraformaldehyde at 18 h postinfection, and immunostained for protein A expression as previously described (31). We determined the number of infected cell clusters identified by protein A expression in ten randomly chosen microscope fields using a 40⫻ objective and calculated total infectious virions on the basis of chamber slide area. Antibodies and inhibitors. Rabbit polyclonal antiserum against FHV protein A has been previously described (31). Rabbit polyclonal antibodies against Hsp60 (SPA-805) and Hsp90 (SPA-846) that cross-react with Drosophila chaperones were from Stressgen Biotechnologies (Victoria, British Columbia, Canada). Rabbit polyclonal antibodies against the influenza virus hemagglutinin (HA) epitope tag were from Santa Cruz Biotechnology (Santa Cruz, CA). Alkaline phosphatase-conjugated anti-digoxigenin Fab fragments were from Roche (Penzberg, Germany), and secondary antibodies for immunoblotting were from Jackson Immunoresearch (West Grove, PA). All inhibitors were from Sigma and stored as concentrated stock solutions at ⫺20°C. Cerulenin and radicicol were dissolved in ethanol at 50 mM and 5 mM, respectively, geldanamycin was dissolved in dimethyl sulfoxide at 5 mM, and lactacystin was dissolved in water at 10 mM. For all inhibitor studies, control cells were treated with comparable solvent concentrations. Plasmids. Standard molecular biology procedures were used for all cloning steps. All Cu2⫹-inducible plasmids were based on the vector pMT-V5/HisA (Invitrogen, Carlsbad, CA), which contains a metallothionein (MT) promoter and simian virus 40 polyadenylation signal to facilitate regulated transcription and translation in Drosophila S2 cells. To generate the Drosophila FHV RNA1 replicon expression plasmid pS2F1, we inserted the ScaI/BsrGI fragment from pF1 (27, 30, 39) into the MscI/Acc65I site of pMT-V5/HisA. To generate the Drosophila protein A expression vector pS2FA, which encodes protein A with a carboxy-terminal HA epitope tag, we first modified pFA-C/HA (30) by placing an encephalomyocarditis virus internal ribosome entry site sequence upstream of

J. VIROL. the protein A open-reading frame and then inserted the Acc65I/XhoI fragment from the resultant intermediate plasmid into the Acc65I/XhoI site of pMT-V5/ HisA. To generate the carboxy-terminal HA-tagged ␤-galactosidase expression plasmid pS2LacZ, we inserted the SpeI/AgeI fragment of pMT-V5/His/LacZ (Invitrogen) into the AvrII/BspEI site of pS2FA. The steroid-inducible rat glucocorticoid receptor expression plasmid pS2GR and the glucocorticoid receptorresponsive luciferase expression plasmid pS2GRE-LUC were generously provided by Jorge Iniguez-Lluhi (University of Michigan). We generated the MT promoter-driven luciferase expression plasmid pS2MT-LUC by inserting the BamHI/SalI fragment of pTRE2/hyg-LUC (BD Biosciences, Palo Alto, CA) into the BamHI/XhoI site of pMT-V5/His/LacZ. S2 cell transfection and induction protocols. S2 cells were cultured in 12-well plates at 106 cells per well and grown in antibiotic-free media for 12 to 18 h prior to lipid-mediated transfection. We used Cellfectin (Invitrogen) and serum- and antibiotic-free media according to the manufacturer’s instructions, and used 2 ␮g expression plasmid and 10 ␮l Cellfectin per well. Cells were incubated for 4 h at 25°C, supplemented with an equal volume of medium containing 10% fetal bovine serum but no antibiotics, and incubated overnight. We induced cells at 24 h after transfection with either 0.5 mM copper sulfate for MT promoterdriven plasmids or 1 ␮M dexamethasone for glucocorticoid receptor-responsive plasmids and harvested cells for RNA or protein expression at 12 h postinduction unless otherwise indicated. For reporter gene assays, we lysed S2 cells in buffer containing 25 mM Tris (pH 7.8), 15 mM MgSO4, 4 mM EGTA, and 1% Triton X-100. We assayed ␤-galactosidase activity with o-nitrophenyl-␤-D-galactopyranoside and quantitated the absorbance at 405 nm. We assayed luciferase activity using a luciferase assay system (Promega, Madison, WI) according to the manufacturer’s instructions. For the generation of stably transfected cells, we cotransfected S2 cells with MT promoter-driven expression plasmids and pCoBlast, a selection plasmid that encodes the blasticidin deaminase gene Bsd under control of the constitutive Drosophila copia promoter (Invitrogen). We initiated selection 48 h after transfection with 25 ␮g blasticidin per ml and routinely observed blasticidin-resistant cells by 1 week in culture. We continued selection for an additional 2 to 3 weeks before stably transfected cells were used for induction, treatment, or RNAi experiments. We used heterogeneous cell populations that had been continuously maintained in culture for less than 3 months for induction and treatment experiments, as protein A expression was unstable after prolonged passage. Immunoblot and Northern blot analyses. Protein samples were prepared and analyzed by immunoblotting as previously described (30, 31) with the following modifications. We used either alkaline phosphatase- or peroxidase-conjugated secondary antibodies and developed immunoblots with Lumi-Phos or SuperSignal West Dura substrates (Pierce, Rockford, IL), respectively. We detected chemiluminescence with an Alpha Innotech Fluorchem 8900 and analyzed images with AlphaEaseFC software. We isolated total RNA from Drosophila S2 cells with TRIzol reagent (Invitrogen), and RNA samples were prepared and analyzed by Northern blotting as previously described (30, 31) with the following modifications. We synthesized strand-specific digoxigenin-labeled riboprobes that corresponded to nucleotides 2718 to 3064 from FHV RNA1 (39) using a Riboprobe in vitro transcription system (Promega) with digoxigenin-11-UTP (Roche) according to the manufacturer’s instructions. Membranes were hybridized with digoxigenin-labeled riboprobes at 62°C and washed extensively with a solution containing 0.1% sodium dodecyl sulfate, 15 mM NaCl, and 1.5 mM sodium citrate, and labeled probes were detected with alkaline phosphataseconjugated antidigoxigenin Fab fragments and chemiluminescence as described above. In vitro and cellular RNA-dependent RNA polymerase (RdRp) assays. A crude replication complex (CRC) membrane fraction containing functional FHV RNA replication complexes was isolated from infected S2 cells at 8 h postinfection, and in vitro RdRp assays were done as previously described (56) with several modifications. Reaction mixtures containing 8 ␮l CRC plus 50 mM Tris (pH 8.0); 50 mM potassium acetate; 15 mM magnesium acetate; 5 ␮g of actinomycin D/ml; 0.5 U of RNAsin/␮l; 1 mM (each) ATP, GTP, and CTP; 50 ␮M UTP; and 5 ␮Ci 3H-labeled UTP in a 25-␮l total volume were incubated for 3 h at 25°C, extracted once with phenol-chloroform, desalted with Bio-Gel P-30 columns (Bio-Rad, Hercules, CA), and separated in 1% nondenaturing agarose gels. After electrophoresis, gels were incubated with Amplify (Amersham Biosciences, Buckinghamshire, England) and 3H-labeled RNA products were detected by fluorography and quantitated by densitometry. For metabolic labeling of viral RNA in cellular RdRp assays, we incubated FHV-infected S2 cells at 8 h postinfection with 5 ␮g actinomycin D/ml for 30 min to block cellular RNA polymerase activity, incubated cells with 20 ␮Ci 3Hlabeled uridine/ml in the presence or absence of inhibitors for 2 h, isolated and

VOL. 79, 2005

Hsp90 AND FHV RNA REPLICATION

6829

FIG. 1. Effect of Hsp90 and fatty acid synthetase inhibitors on S2 cell growth, viability, and infectious virion production. (A) Uninfected S2 cells were incubated with vehicle only (circles), 50 ␮M cerulenin (triangles), 5 ␮M geldanamycin (squares), or 5 ␮M radicicol (diamonds), and cell viability (closed symbols, left y axis) and cell number (open symbols, right y axis) were determined by trypan blue exclusion with a hemocytometer. Individual data points represent the mean values from three experiments. (B) S2 cell were infected with FHV at an MOI ⫽ 10 and simultaneously treated with vehicle only (None), 50 ␮M cerulenin (Cer), 5 ␮M geldanamycin (GA), or 5 ␮M radicicol (Rad). Infectious virions were harvested at 12 h postinfection, purified by pelleting through sucrose gradients to remove residual inhibitors, and quantitated by an immunofluorescencebased assay. Results represent the means ⫾ standard errors of the means (SEM) from three experiments and are expressed as severalfold increases over the calculated virus input at an MOI ⫽ 10 to account for inoculum virion recovery. FHV virion production in infected S2 cells is extremely vigorous, with an estimated burst size of approximately 250,000 particles per cell (47).

separated total RNA in denaturing 1.4% agarose–formaldehyde gels, and detected and quantitated 3H-labeled RNA product as described above. Double-stranded RNA (dsRNA) production and RNAi protocol. We amplified approximately 700-bp DNA fragments from the 5⬘ coding region of the lacZ and Drosophila Hsp83 genes that included the initiator ATG codon by using PCR and primer pairs that incorporated a 5⬘ T7 RNA polymerase binding site, which are the underlined regions in the primer sequences below. For lacZ amplification, we used pMT-V5/His/LacZ as a template, the sense primer 5⬘-TAATACGACTCA CTATAGGGATTCTGCAGATCGAAACGA-3⬘, and the antisense primer 5⬘TAATACGACTCACTATAGGGAAAGCGAGTGGCAACATGG-3⬘. For Hsp83 amplification, we used Drosophila cDNA that was generated from total S2 cellular RNA by reverse transcription with oligo(dT) primers as a template, the sense primer 5⬘-TAATACGACTCACTATAGGGTTTGTAAATCCATTGCA GA-3⬘, and the antisense primer 5⬘-TAATACGACTCACTATAGGGCTCA TCAGTCTCCATCTCC-3⬘. We purified PCR products using a Wizard PCR Clean-up system (Promega) and generated RNA by in vitro transcription using an Ampliscribe T7 High Yield kit (Epicentre Technologies, Madison, WI). The cRNA products were ethanol precipitated, resuspended in water, heated to 65°C for 30 min, and cooled slowly at room temperature to allow dsRNA complexes to form. Final dsRNA products were quantitated by spectrophotometry and analyzed by nondenaturing agarose gel electrophoresis to ensure that the majority of the dsRNA existed as a single band of approximately 700 bp (data not shown). We used RNAi conditions for S2 cells as described by Clemens et al. (11) with the following modifications. We diluted stably transfected S2 cells to 106 cells per ml in serum-free Schneider’s medium, dispensed 350 ␮l per well into 12-well tissue culture plates, and added 10 ␮g dsRNA per well. Plates were swirled briefly to disperse dsRNA and incubated at 25°C for 1 h, 650 ␮l medium with 10% fetal bovine serum was added per well, and cells were cultured for an additional 48 h to allow maximal depletion of target proteins prior to induction.

RESULTS Hsp90 inhibition suppresses the production of infectious virions and the accumulation of viral RNA and protein A in FHV-infected Drosophila S2 cells. On the basis of the previously identified role of cellular chaperones in viral replication (49), we hypothesized that FHV may also use cellular chaperones to facilitate viral RNA replication. In addition, genomewide microarray analyses of FHV-infected S2 cells revealed

the transcriptional upregulation of several Drosophila genes encoding Hsp90 chaperone complex components (D. Miller and P. Ahlquist, unpublished data). Thus, we initiated studies to elucidate the role of the Hsp90 chaperone complex in FHV replication in Drosophila cells. To determine the functional impact of Hsp90 activity on FHV replication, we examined the effects of Hsp90 inhibitors on virus-infected Drosophila S2 cells. Geldanamycin, a benzoquinone ansamycin, and radicicol, a structurally unrelated antibiotic, are potent and specific inhibitors of Hsp90 activity in vivo and in vitro (34, 42, 45). We used cerulenin, a fatty acid synthetase inhibitor previously shown to suppress positivestrand RNA virus replication (16, 36), as a positive control. We chose initial inhibitor concentrations of 50 ␮M cerulenin, 5 ␮M geldanamycin, and 5 ␮M radicicol on the basis of published studies of cultured cells that maximized target inhibition and minimized cellular toxicity (16, 20, 23, 36, 54). None of the inhibitors significantly reduced S2 cell viability during a 24-h incubation, although both geldanamycin and cerulenin reduced cell proliferation such that by 24 h cell recovery was 50 to 60% of control levels (Fig. 1A), and thus we routinely analyzed cells at 12 h postinfection unless otherwise indicated. We initially examined whether Hsp90 inhibition suppressed infectious virion production (Fig. 1B). Cells were infected with FHV at an MOI of 10 and simultaneously treated with 50 ␮M cerulenin, 5 ␮M geldanamycin, or 5 ␮M radicicol. Infectious virions were harvested at 12 h postinfection, purified by pelleting through sucrose gradients to remove residual inhibitors, and quantitated by an immunofluorescence-based infection assay. Compared to controls treated with vehicle only, the Hsp90 inhibitors geldanamycin and radicicol reduced infectious virion production by more than 2 logs, in similarity to the reduction seen with the fatty acid synthetase inhibitor cerulenin (Fig. 1B). These results suggested that infectious virion production

6830

KAMPMUELLER AND MILLER

required, at least in part, a functional Hsp90 chaperone complex. The presence of infectious virions represents the culmination of multiple steps in the viral life cycle. To determine the effects of Hsp90 inhibition on viral replication steps prior to genome encapsidation, maturation, and virion release, we examined FHV RNA and protein A accumulation by quantitative Northern blotting and immunoblotting in infected S2 cells treated with geldanamycin or radicicol (Fig. 2). Protein A is the essential viral protein required for FHV RNA replication complex assembly, genomic positive-strand RNA1 [(⫹)RNA1] and subgenomic (⫹)RNA3 are primary products of RNA replication complex activity, and negative-strand RNA1 [(⫺)RNA1] is the template for genomic (⫹)RNA1 synthesis (4). We observed a significant reduction in the accumulation of protein A, (⫹)RNA1, (⫺)RNA1, and (⫹)RNA3 in FHV-infected S2 cells treated with cerulenin (Fig. 2A, lane 3), geldanamycin (Fig. 2A, lane 4), or radicicol (Fig. 2A, lane 5) when inhibitors were added at the time of infection. When compared quantitatively to control infected but untreated cells, the Hsp90 inhibitor geldanamycin reduced protein A, (⫹)RNA1, (⫺)RNA1, and (⫹)RNA3 accumulation by 95%, 98%, 87%, and 95%, respectively (Fig. 2B). Similar quantitative results were obtained with Hsp90 inhibitor radicicol and to a lesser degree with the fatty acid synthetase inhibitor cerulenin. To further examine the potency of geldanamycin with respect to its suppression of FHV RNA and protein A accumulation, we performed titration experiments (Fig. 2C). We treated infected S2 cells at the time of infection with increasing concentrations of geldanamycin and quantitatively analyzed the accumulation of FHV protein A, (⫹)RNA1, (⫺)RNA1, and (⫹)RNA3 at 12 h postinfection. Geldanamycin reduced FHV protein A and RNA accumulation by 50% at 14 nM, a 50% inhibitory concentration (IC50) consistent with previous studies of the inhibitory effects of geldanamycin on hepatitis C virus protease activity (54) and vaccinia virus growth (23) in cultured cells. These results indicated that pharmacological inhibition of Hsp90 activity potently suppressed FHV RNA and protein A accumulation in infected Drosophila S2 cells and suggested that Hsp90 was involved in FHV RNA replication complex assembly or function. We also examined whether delayed treatment experiments altered the suppressive effects of Hsp90 inhibitors on FHV RNA and protein A accumulation. We infected S2 cells as described above but added inhibitors at 4 h postinfection, when functional viral RNA replication complexes had already been assembled (31). Under these treatment conditions, geldanamycin (Fig. 2A, lane 6) and radicicol (Fig. 2A, lane 7) also reduced protein A and viral RNA accumulation compared to untreated control results (Fig. 2A, lane 2, and Fig. 2B). To take into account the estimated levels of preexisting functional viral RNA replication complexes present at 4 h postinfection, we compared (⫺)RNA1:(⫹)RNA1 and protein A:(⫹)RNA1 ratios (Table 1). Template (⫺)RNA1 and protein A are essential components of FHV RNA replication complexes, whereas (⫹)RNA1 is a major product of viral RNA replication complex activity. We calculated baseline (⫺)RNA1:(⫹)RNA1 and protein A:(⫹)RNA1 ratios from infected but untreated cells by dividing the percentages of (⫺)RNA1, protein A, or (⫹)RNA1 present at 4 h postinfection relative to 12 h postinfection. If

J. VIROL.

FIG. 2. Hsp90 inhibition suppresses FHV RNA replication in infected cells. (A) S2 cells were infected with FHV and treated with 50 ␮M cerulenin (C), 5 ␮M geldanamycin (G), or 5 ␮M radicicol (R) at 0 h postinfection (lanes 3 to 5) or 4 h postinfection (lanes 6 and 7) and harvested at 12 h postinfection. We used cerulenin as a positive control, as this fatty acid synthetase inhibitor has been previously shown to suppress positive-strand RNA virus replication (16, 36). Total protein or RNA from an equal number of cells was separated by electrophoresis, and protein A (upper blot) was detected by immunoblotting, while (⫹)RNA1 and (⫹)RNA3 (third blot) or (⫺)RNA1 (fourth blot) was detected by Northern blotting. The Hsp60 immunoblot and rRNA band detected by ethidium bromide staining are shown as loading controls. (B) Quantitative data for FHV protein A, genomic (⫹)RNA1, template (⫺)RNA1, and subgenomic (⫹)RNA3 accumulation are relative to the control (Fig. 2A, lane 2) and represent the means ⫾ SEM from three experiments. (C) Titration of geldanamycin (GA) effect on FHV protein A and RNA accumulation. Cells were infected and treated with increasing GA concentrations at the time of infection, and viral protein A and RNA accumulation were analyzed at 12 h postinfection as described above. Individual data points represent the mean values from three experiments.

VOL. 79, 2005

Hsp90 AND FHV RNA REPLICATION

6831

TABLE 1. FHV RNA and protein A ratios in S2 cells treated with Hsp90 inhibitors 4 h after infection % Relative accumulationa Treatment

Noneb Geldanamycinc Radicicolc

Ratio

(⫹)RNA1

(⫺)RNA1

Ptn A

(⫺)RNA1: (⫹)RNA1

Ptn A: (⫹)RNA1

6.9 44.4 47.6

16.4 41.2 48.8

48.0 56.1 75.9

2.4 0.9 1.0

7.0 1.3 1.6

a Values represent the percentage of RNA or protein A (Ptn A) accumulation relative to infected but untreated cells harvested at 12 h postinfection. b Results are from infected but untreated cells harvested at 4 h postinfection. c Results are from infected cells treated at 4 h postinfection and harvested at 12 h postinfection and are taken from the data represented in Fig. 2B.

Hsp90 inhibitors had equivalent suppressive effects on both the assembly and function of FHV RNA replication complexes, the (⫺)RNA1:(⫹)RNA1 and protein A:(⫹)RNA1 ratios from infected cells treated with geldanamycin or radicicol at 4 h postinfection and harvested at 12 h postinfection should be similar to baseline ratios, assuming no significant differences in FHV RNA or preformed functional RNA replication complex stability. In contrast, these ratios would decrease if Hsp90 inhibitors predominantly affected viral RNA replication complex assembly or would increase if the predominant effect was on replication complex function. Treatment with either geldanamycin or radicicol at 4 h postinfection reduced both (⫺)RNA1:(⫹)RNA1 and protein A:(⫹)RNA1 ratios (Table 1). These results suggested that the suppressive effects of Hsp90 inhibitors were predominantly on the assembly of functional FHV RNA replication complexes. Hsp90 inhibition does not suppress FHV RdRp activity in vitro or in cells. Hsp90 is a cytosolic protein (38), whereas FHV RNA replication complexes localize to mitochondrial membranes (31). Initial cellular fractionation studies demonstrated that Hsp83, the Drosophila Hsp90-family chaperone (12), did not repartition into the membrane fraction and cosediment with protein A in lysates from FHV-infected S2 cells, suggesting that Hsp90 chaperones were not tightly bound to functional membrane-associated FHV RNA replication complexes (data not shown). However, to specifically examine whether Hsp90 activity was important for preformed FHV RNA replication complex function, we analyzed the effects of geldanamycin on FHV RdRp activity in vitro and in cells (Fig. 3). Membrane fractions from FHV-infected Drosophila cells contain functional viral RNA replication complexes whose RdRp activity can be measured in vitro (56). We isolated and incubated membrane fractions from uninfected and FHV-infected S2 cells with 3H-labeled UTP and analyzed the synthesis of viral dsRNA products by nondenaturing agarose gel electrophoresis and fluorography (Fig. 3A). No exogenous FHV protein A or RNA templates were added to the RdRp reactions, and thus any in vitro RNA synthesis represented functional preformed viral RNA replication complex activity. We observed no significant suppression of FHV RdRp activity in vitro with up to 50 ␮M geldanamycin (Fig. 3A, lane 4), a concentration well above levels previously shown to suppress Hsp90 activity in vitro (20, 54, 57), 10-fold higher than that used in cellular assays (Fig. 1 and 2), and approximately 3,500fold above the IC50 for geldanamycin and FHV RNA replica-

FIG. 3. Hsp90 inhibition does not suppress FHV RdRp activity in vitro or in cells. (A) Membrane fractions from control S2 cells (lane 1) or FHV-infected S2 cells that contain a crude replication complex (CRC) (lanes 2 to 6) were incubated with 3H-labeled UTP and unlabeled ribonucleotides and vehicle only (lane 2), 5 mM EDTA (lane 3), or decreasing concentrations of geldanamycin (GA; lanes 4 to 6), and the reaction products were separated by nondenaturing agarose gel electrophoresis and detected by fluorography. The positions of RNA molecular weight (MW) markers transcribed in vitro are shown on the left. The major radiolabeled bands corresponding to probable FHV dsRNA1 (⬃6,200 nucleotides) and dsRNA2 (⬃2,800 nucleotides), the predominant FHV RdRp in vitro products (56), were quantitated by densitometry, and the numbers represent the mean values from two experiments relative to untreated controls (lane 2). (B) Total RNA from control (lane 1) or FHV-infected (lanes 3 to 5) S2 cells incubated with 3H-labeled uridine for 2 h in the presence of actinomycin D and vehicle only (lane 3), 5 mM EDTA (lane 4), or 5 ␮M geldanamycin (GA; lane 5) was separated by denaturing agarose gel electrophoresis and analyzed by fluorography. The positions of all three FHV RNA species are shown on the right. 3H-labeled RNA from S2 cells expressing an FHV RNA1 replicon (lane 2) (Fig. 4A) is shown as reference for the identification of RNA1 and RNA3 bands from infected cells, and the rRNA band detected by ethidium bromide staining is shown as a loading control. There is a faint background radiolabeled band visible in both mock-infected cells (lane 1) and cells expressing an FHV RNA1 replicon (lane 2) that comigrates with RNA2. Results are from one representative experiment of two.

tion in infected cells (Fig. 2C). In contrast, EDTA efficiently suppressed viral RNA synthesis (Fig. 3, lane 3), a result consistent with the previously identified dependence of nodavirus RdRp activity on divalent cations (15). We also examined the impact of Hsp90 activity on FHV RNA replication complex activity in infected cells using metabolic labeling (Fig. 3B). We incubated S2 cells at 8 h postinfection with actinomycin D, which inhibits cellular RNA polymerase but not FHV RdRp activity (24), and subsequently labeled viral RNA by incubation with 3H-labeled uridine for 2 h in the presence or absence of 5 ␮M geldanamycin, a

6832

KAMPMUELLER AND MILLER

concentration that efficiently suppresses Hsp90 chaperone complex activity in Drosophila cells (2) (Fig. 4C) and potently suppressed FHV RNA accumulation in infected cells (Fig. 2). Consistent with the in vitro RdRp results (Fig. 3A), geldanamycin did not inhibit viral RNA synthesis in FHV-infected cells (Fig. 3B, lane 5), whereas EDTA significantly reduced the accumulation of 3H-labeled FHV RNA (Fig. 3B, lane 4). These results were in marked contrast to the suppressive effects of geldanamycin on FHV RNA accumulation in infected cells (Fig. 2) and suggested that Hsp90 activity was not essential for preformed FHV RNA replication complex function. Hsp90 inhibition suppresses plasmid-directed FHV RNA replication in S2 cells. The suppressive effects of geldanamycin and radicicol on viral RNA and protein A accumulation in FHV-infected S2 cells (Fig. 2) may have been due to disruption of an Hsp90-dependent process related to viral attachment, entry, or uncoating. To specifically examine viral RNA replication in the absence of infectious virions, we developed a plasmid-based replicon system for FHV RNA1 replication and protein A expression in Drosophila S2 cells (Fig. 4A) that was similar to the system used to study FHV RNA replication in S. cerevisiae (27, 30, 32, 39). The Drosophila expression system (Invitrogen) is designed for inducible metallothionein (MT) promoter-driven gene expression in either transiently or stably transfected S2 cells. Transient transfection of S2 cells with the wild-type RNA1 expression plasmid pS2F1 resulted in the accumulation of protein A, (⫹)RNA1, and (⫹)RNA3 after Cu2⫹ induction (Fig. 4B, lane 2). We induced with 0.5 mM Cu2⫹ to prevent the heavy metal-induced heat shock response previously described for Cu2⫹ concentrations greater than 1 mM in cultured Drosophila cells (7). Cellular fractionation and confocal immunofluorescence microscopy studies indicated that protein A was membrane associated and localized to mitochondria in pS2F1-transfected cells (data not shown) similarly to FHV-infected Drosophila cells (31) and yeast expressing an FHV RNA1 replicon (30). To examine whether Hsp90 inhibition suppressed viral RNA replication in S2 cells expressing an FHV RNA1 replicon, we transiently transfected cells with pS2F1 or empty vector, induced with Cu2⫹ in the presence of 50 ␮M cerulenin, 5 ␮M geldanamycin, or 5 ␮M radicicol, and analyzed FHV RNA and protein A expression 12 h postinduction (Fig. 4B). In similarity to findings obtained with FHV-infected S2 cells (Fig. 2), we observed a reduction in the accumulation of protein A, (⫹)RNA1, and (⫹)RNA3 in pS2F1-transfected S2 cells treated with cerulenin (Fig. 4B, lane 3), geldanamycin (Fig. 4B, lane 4), or radicicol (Fig. 4B, lane 5). The approximately 90% reduction in protein A accumulation in pS2F1-transfected cells treated with Hsp90 inhibitors was consistent with the quantitative results obtained with FHV-infected S2 cells (Fig. 2B). In contrast, the 60 to 70% reduction in subgenomic (⫹)RNA3 accumulation was less than that observed in infected cells (Fig. 2B) and may have been related to the increase in (⫹)RNA3 accumulation relative to (⫹)RNA1 in cells expressing FHV RNA1 replicons compared to FHV-infected cells (compare Fig. 4B, lane 2, with Fig. 2A, lane 2). An increased (⫹)RNA3: (⫹)RNA1 ratio relative to that observed in virus-infected cells has previously been observed in both yeast and Drosophila DL-1 cells expressing FHV RNA1 replicons (27, 39, 40) and is likely related to the absence of RNA2-mediated suppression of

J. VIROL.

subgenomic RNA3 synthesis (4). An alternative explanation for the apparent reduced suppressive effects of Hsp90 inhibitors on FHV RNA replication in pS2F1-transfected cells was leaky MT promoter activity and the formation of functional viral RNA replication complexes before Hsp90 inhibitor addition, as we induced cells in the presence of inhibitors 24 h after transient transfection to reduce any potential confounding effects of Hsp90 inhibition on transfection efficiency. We examined whether the suppressive effects of Hsp90 inhibition on FHV RNA replication in pS2F1-transfected cells was due to suppression of MT promoter activity or to a general suppression of plasmid-directed transcription or translation (Fig. 4C). As a positive control, we used a luciferase reporter gene under the control of an Hsp90-dependent glucocorticoid receptor-responsive promoter (37) and found that 5 ␮M geldanamycin reduced dexamethasone-induced luciferase activity by 90% (Fig. 4C, left bars). In S2 cells transfected with a plasmid encoding an MT promoter-driven luciferase reporter gene, geldanamycin also suppressed Cu2⫹-induced luciferase activity, although to a lesser degree than in the positive control (Fig. 4C, middle bars). However, the moderate suppression of luciferase activity by geldanamycin may have been due to a direct effect on reporter gene activity rather than MT promoter suppression, as Hsp90 can stabilize luciferase in vitro (51). Consistent with this hypothesis, geldanamycin did not suppress MT promoter-driven ␤-galactosidase activity (Fig. 4C, right bars). These results indicated that most, if not all, of the inhibitory effects of geldanamycin in pS2F1-transfected cells (Fig. 4B) were due to specific suppression of viral RNA and protein A accumulation. Thus, together with the in vitro and cellular RdRp activity results (Fig. 3), the results with FHV RNA1 replicons suggested that Hsp90 functioned at a step in viral RNA replication between the initial release of virion RNA and the assembly of functional viral RNA replication complexes. Hsp90 inhibition suppresses FHV protein A accumulation in the absence of viral RNA replication. Protein A is the essential viral protein for FHV RNA replication complex assembly and function and is a potential target for the suppressive effects of Hsp90 inhibitors. Thus, we examined whether Hsp90 inhibition reduced protein A accumulation in the absence of viral RNA replication. To better control transfection efficiency, eliminate potential confounding effects of transient transfection on Hsp90 expression and activity, and facilitate the RNAi studies described below, we generated S2 cells stably transfected with an MT promoter-driven protein A expression plasmid. This plasmid, pS2FA, resembles the yeast protein A expression plasmid pFA (27, 30, 32, 39), where the FHV RNA1 5⬘ and 3⬘ nontranslated regions were modified to improve its translation efficiency but disrupt its function as a replication template. We also incorporated a carboxy-terminal HA epitope tag into pS2FA and a control MT promoter-driven ␤-galactosidase expression vector, pS2LacZ, to facilitate immunodetection. Carboxy-terminal epitope tags do not disrupt protein A expression, intracellular localization, or RdRp function (30, 32). We cotransfected S2 cells with pS2FA or pS2LacZ and the selection plasmid pCoBlast and grew cells for at least 3 weeks in the presence of blasticidin to select resistant cells. Preliminary experiments with S2 cells stably transfected with pS2FA showed the accumulation of protein A but not

VOL. 79, 2005

Hsp90 AND FHV RNA REPLICATION

6833

FIG. 5. Hsp90 inhibition suppresses FHV protein A accumulation in the absence of viral RNA replication. (A) S2 cells stably transfected with pS2FA (upper two blots) or pS2LacZ (bottom blot) were induced with 0.5 mM Cu2⫹ (lanes 2 to 7) in the absence (lane 2) or presence (lanes 3 to 7) of decreasing geldanamycin concentrations (GA conc.). The relative height of the triangle represents geldanamycin concentrations, which are 10-fold dilutions between adjacent lanes. Total protein from an equivalent number of cells was harvested at 12 h postinduction and analyzed for protein A (Ptn A) or ␤-galactosidase (␤-Gal) accumulation by anti-HA immunoblotting. The Hsp60 immunoblot is shown as a loading control. A background band detected by the polyclonal anti-HA antibody is present that comigrates with ␤-galactosidase (120 kDa) but above protein A (112 kDa). (B) S2 cells transfected with pS2FA (lanes 2 to 5) were induced in the presence of 1 ␮M GA (lane 3) or GA plus the proteasome inhibitor lactacystin (Lact; lanes 4 and 5), harvested at 12 h postinduction, and immunoblotted for protein A (upper blot) or Hsp60 (lower blot). The numbers represent the mean protein A levels from two or three experiments relative to control (lane 2).

FIG. 4. Hsp90 inhibition suppresses FHV RNA replication in the absence of infectious virions. (A) Schematic of pS2F1, an MT promoter-driven plasmid that expresses an FHV RNA1 replicon. Authentic viral 5⬘ and 3⬘ RNA termini were generated in cells through precise placement of the MT promoter start site and a hepatitis ␦ ribozyme (Rz), respectively. Drosophila RNA polymerase II-transcribed FHV RNA1 can serve as a template for both replication and translation into protein A, and thus viral RNA replication and subgenomic (sg) RNA3 synthesis become autonomous and MT promoter independent after Cu2⫹ induction. (B) S2 cells transiently transfected with empty vector (lane 1) or pS2F1 (lanes 2 to 5) were induced with 0.5 mM Cu2⫹ in the presence of 50 ␮M cerulenin (C; lane 3), 5 ␮M geldanamycin (G; lane 4), or 5 ␮M radicicol (R; lane 5) 24 h after transfection. To control for transfection efficiency, we transfected cells with pS2F1 as a single culture and subsequently aliquoted cells into 12-well tissue culture plates for induction and treatment with individual inhibitors. Total protein and RNA were harvested from an equal number of cells at 12 h postinduction and analyzed by immunoblotting and Northern blotting. The numbers represent the levels of protein A and (⫹)RNA3 accumulation relative to control (lane 2) and are the mean values from

template (⫺)RNA1 or subgenomic (⫹)RNA3 and also revealed that protein A was membrane associated and localized to mitochondria (data not shown), in similarity to results obtained with yeast transformed with pFA (27, 39). Titration experiments demonstrated that protein A accumulation in S2 cells stably transfected with pS2FA peaked at 24 h after induc-

three experiments. The potent suppression of protein A accumulation by the fatty acid synthetase inhibitor cerulenin in pS2F1-transfected cells (lane 3) was in contrast to its moderate suppression of protein A accumulation in FHV-infected cells (Fig. 2A, lane 3) and may reflect a differential effect of cerulenin on the translation of FHV RNA1 derived from cytosolic virions versus plasmid-derived nuclear transcripts. (C) Metallothionein (MT) promoter- or glucocorticoid receptor responsive (GRE) promoter-driven luciferase (LUC) or ␤-galactosidase (LacZ) activity in transiently transfected S2 cells in the absence (black bars) or presence (gray bars) of 5 ␮M geldanamycin (GA). To control for transfection efficiency, cells were transfected with individual reporter plasmids as a single culture as described above. Results are the means ⫾ SEM from three experiments and are expressed as percentages of control reporter gene activity, where activity in the absence of geldanamycin was set at 100%.

6834

KAMPMUELLER AND MILLER

tion with 0.5 mM Cu2⫹ (data not shown), but due to the inhibitory effects of geldanamycin on cell proliferation by 24 h in culture (Fig. 1B), we analyzed pS2FA-transfected cells at 12 h postinduction similarly to experiments with FHV-infected (Fig. 2) and pS2F1-transfected (Fig. 4) cells. We initially examined the impact of Hsp90 inhibition with geldanamycin on protein A or ␤-galactosidase accumulation in cells stably transfected with pS2FA or pS2LacZ, respectively (Fig. 5A). Geldanamycin reduced Cu2⫹-induced protein A accumulation in a dose-dependent manner (Fig. 5A, top blot) but had no effect on ␤-galactosidase accumulation (Fig. 5A, bottom blot). The estimated IC50 for geldanamycin and protein A accumulation in pS2FA-transfected cells was 40 nM, similar to the IC50 for geldanamycin and the accumulation of protein A and RNA in FHV-infected S2 cells (Fig. 2C). We consistently observed residual protein A accumulation even with high geldanamycin concentrations (Fig. 5A, top blot, lanes 3 to 5), suggesting that either the concentrations of inhibitor used in these experiments did not completely suppress Hsp90 activity or that the accumulation of a small fraction of protein A was Hsp90 independent. We did not detect Cu2⫹-independent protein A expression in cells stably transfected with pS2FA (Fig. 5A, top blot, lane 1), suggesting that residual protein A accumulation was not due to leaky MT promoter activity. Several cellular proteins whose maturation and function depend on Hsp90 activity are retargeted to the proteasome for degradation in the presence of Hsp90 inhibitors (5, 28, 46, 55). Thus, we examined protein A accumulation in pS2FA-transfected cells treated with both geldanamycin and the proteasome inhibitor lactacystin (25) at the time of induction (Fig. 5B). Lactacystin partially attenuated the suppressive effect of geldanamycin on protein A accumulation in a dose-dependent manner (Fig. 5B, lanes 4 and 5). These results indicated that Hsp90 activity was important for protein A accumulation in the absence of viral RNA replication and suggested that the impact of Hsp90 inhibition on FHV RNA accumulation (Fig. 2 and 4) was due in part to suppression of replication complex assembly, potentially via premature proteasome-mediated degradation of protein A. RNAi-mediated Hsp83 downregulation suppresses FHV protein A accumulation. The specificity of geldanamycin and radicicol for Hsp90 (34, 42, 45) facilitates the study of Hsp90dependent cellular pathways. However, to independently confirm the role of Hsp90 in FHV protein A accumulation in S2 cells using a genetic approach that did not rely on pharmacological inhibitors, we used RNAi to selectively downregulate the expression of the Drosophila Hsp90-family chaperone Hsp83 (12) (Fig. 6). Drosophila cells are particularly susceptible to RNAi (9), and this approach has been previously used to downregulate Hsp83 expression in S2 cells (29), despite its essential function and relatively high abundance in the cytosol (38). We generated 700-bp dsRNA products corresponding to the 5⬘ coding sequence of Hsp83 or control lacZ, which we used in RNAi experiments with S2 cells stably transfected with pS2LacZ or pS2FA. FHV subgenomic RNA3 and hence protein B are not produced in S2 cells transfected with pS2FA, and thus we avoided the potential confounding effects of protein B-mediated RNAi suppression (26). Initial experiments demonstrated that 48 h was the optimal time period for maximal RNAi-mediated Hsp83 suppression (data not shown),

J. VIROL.

FIG. 6. RNAi-mediated suppression of Hsp83 in Drosophila cells reduces protein A accumulation. (A) S2 cells stably transfected with pS2LacZ were treated with 10 ␮g lacZ (lane 3) or Hsp83 (lane 4) dsRNA. After 48 h, cells were induced with 0.5 mM Cu2⫹ in the presence (lane 2) or absence (lanes 1, 3, and 4) of 1 ␮M geldanamycin (GA) and harvested at 12 h postinduction, and lysates from an equal number of cells were analyzed for ␤-galactosidase (␤-Gal), Hsp83, and Hsp60 accumulation by immunoblotting. (B) S2 cells stably transfected with pS2FA were treated with 10 ␮g lacZ dsRNA (lane 3) or decreasing amounts of Hsp83 dsRNA (lanes 4 to 7), induced, harvested, and analyzed as described above. The relative height of the triangle represents the amount of Hsp83 dsRNA added to cultures, which are twofold dilutions between adjacent lanes. (C) Quantitative data for Hsp83 (black bars) and protein A (gray bars) accumulation in pS2FAtransfected cells. Results are the mean values from two or three experiments relative to untreated controls (Fig. 6B, lane 1).

VOL. 79, 2005

Hsp90 AND FHV RNA REPLICATION

consistent with previous observations (29). Cells stably transfected with pS2LacZ and treated with Hsp83 dsRNA showed a specific 70 to 80% reduction in Hsp83 accumulation (Fig. 6A, lane 4), whereas treatment with control lacZ dsRNA suppressed only ␤-galactosidase expression after induction (Fig. 6A, lane 3). In cells stably transfected with pS2FA, Hsp83 dsRNA suppressed protein A accumulation in a dose-dependent manner (Fig. 6B, lanes 4 to 7), whereas lacZ dsRNA had minimal effect (Fig. 6B, lane 3). When examined quantitatively, there was a significant correlation (R ⫽ 0.796; P ⬍ 0.002) between Hsp83 and protein A accumulation (Fig. 6C). These results supported the Hsp90 inhibitor studies (Fig. 5) and confirmed the role of Hsp90 chaperones in FHV protein A accumulation in Drosophila cells. DISCUSSION In this study, we investigated the role of the molecular chaperone Hsp90 in FHV RNA replication in Drosophila S2 cells. On the basis of these experiments, we drew three main conclusions. First, pharmacological inhibitors of Hsp90 activity potently suppressed the production of infectious virions, and the accumulation of viral RNA and protein A in FHV-infected cells, but did not significantly affect the activity of preformed FHV RNA replication complexes in vitro or in cells. Second, Hsp90 inhibition suppressed viral RNA and protein A accumulation in cells expressing an FHV RNA1 replicon. Third, either pharmacologic inhibition of Hsp90 activity or RNAimediated chaperone downregulation suppressed FHV protein A accumulation in the absence of viral RNA replication. These observations identify Hsp90 as a cellular host factor important for FHV replication in Drosophila cells and suggest that molecular chaperones participate in the assembly of FHV RNA replication complexes. The identification of Hsp90 as a host factor involved in FHV RNA replication is consistent with previous studies on the impact of this cellular chaperone on viral replication and pathogenesis. In particular, Hsp90 has been shown to facilitate vaccinia virus replication (23), hepatitis C virus protease maturation (54), influenza virus RNA synthesis (33), and hepatitis B virus (HBV) reverse transcriptase activity (20–22). However, Hsp90 appears to participate at different stages in the viral life cycle of specific viruses. Geldanamycin inhibits vaccinia virus growth, DNA replication, and intermediate gene transcription in cultured cells through an undefined mechanism, although the association of Hsp90 and the vaccinia virus 4a core protein suggests a role in either viral core uncoating or intracellular transport of incoming cores (23). For hepatitis C virus, Hsp90 physically interacts with the nonstructural 2/3 (NS2/3) protease, and Hsp90 inhibitors disrupt the NS2/3-mediated proteolytic cleavage necessary for NS3 release (54). However, the significance of this observation in the context of hepatitis C virus replication within cells is unknown. Similarly, although Hsp90 physically interacts with influenza virus RNA polymerase and stimulates its activity in vitro (33), the impact of Hsp90 on influenza virus replication in infected cells in also unknown. In contrast, Hsp90 has been shown to facilitate HBV replication within cells (20), and the mechanism whereby Hsp90 impacts HBV replication has been studied in detail. Hsp90 physically mediates the interactions between the HBV reverse

6835

transcriptase and a specific pregenomic viral RNA signal, a step that is essential for the protein-primed initiation of reverse transcription and the assembly of replication-competent nucleocapsids (20, 22). A similar mechanism is possible with FHV, where Hsp90 could mediate the interactions between protein A and a viral RNA template to facilitate replication complex assembly. However, Hsp90 inhibition suppressed protein A accumulation in the absence of viral RNA replication (Fig. 5 and 6), suggesting that Hsp90 could facilitate a template-independent step in FHV RNA replication complex assembly. The observations that proteasome inhibition partially attenuated the suppressive effect of geldanamycin on protein A accumulation (Fig. 5B), and that preformed FHV RNA replication complex activity was not geldanamycin sensitive (Fig. 3), suggest that Hsp90 activity is important for protein A stability during replication complex assembly. Preliminary results indicate that protein A expressed in the absence of viral RNA replication and Hsp90 inhibition was stable for at least 20 h in S2 cells (K. Stapleford and D. Miller, unpublished data), consistent with previous observations of nodavirus RNA polymerase stability in infected cells (13). Since the predominant fraction of protein A is membrane associated within cells (31), protein A stability may depend on Hsp90 chaperone complex activity prior to membrane association. However, our results do not exclude a role for Hsp90 in FHV RNA replication complex turnover, although the cellular RdRp activity results suggested that geldanamycin did not promote the rapid degradation of functional viral RNA replication complexes (Fig. 3B). Studies are in progress to specifically examine the impact of Hsp90 on defined stages of FHV RNA replication complex assembly, such as protein A synthesis, degradation, intracellular trafficking, and membrane association. The Hsp90 chaperone complex normally mediates the maturation of a select group of cellular proteins, termed client proteins, many of which function as key regulators of cell growth, differentiation, and death (38). The best-studied examples of Hsp90-client proteins are the steroid receptors, where the Hsp90 chaperone complex maintains cytosolic receptors in a ligand-accessible conformation (38). In addition, Hsp90 has been shown to facilitate the maturation and intracellular targeting of several membrane proteins, including the Src-kinase p56lck (5), the epidermal growth factor receptor (50), and the cystic fibrosis transmembrane conductance regulator (28). The intrinsic ATPase activity of Hsp90 is essential for its chaperone function (35), and the Hsp90-specific inhibitors geldanamycin and radicicol bind to the amino-terminal ATP/ADP-binding domain of Hsp90 to disrupt its enzymatic activity (42). In the presence of pharmacological inhibitors, Hsp90 client proteins are redirected to the proteasome for degradation (5, 28, 46, 55). Our results are consistent with a model whereby Hsp90 facilitates FHV RNA replication complex assembly via a direct but transient physical interaction with protein A, in similarity to other client proteins (38). However, we cannot exclude an indirect mechanism whereby Hsp90 inhibition disrupts the maturation or function of a cellular protein necessary for FHV RNA replication complex assembly in Drosophila cells. Mitochondrial proteins are inserted into the outer membrane or imported into the intermembrane space or matrix posttranslationally (53), in contrast to the cotranslational in-

6836

KAMPMUELLER AND MILLER

sertion and import of many proteins destined for the endoplasmic reticulum (41). The cytosolic stabilization and intracellular targeting of mitochondrial proteins is mediated by several chaperones (19), including mitochondrial-specific chaperones such as Mtf52 (10) and mitochondrial import stimulation factor (17), and chaperones with more general membrane-targeting functions such as the nascent polypeptide-associated complex (14) and members of the Hsp40 (8) and Hsp70 (48) families. In addition, Young et al. recently demonstrated that Hsp90 can also deliver mitochondrial preproteins to the mitochondrial membrane in mammalian cells via interactions with the import complex receptors, whereas this function is mediated by Hsp70 in yeast (57). FHV protein A and viral RNA replication complexes localize to mitochondrial outer membranes in both Drosophila (31) and S. cerevisiae (30) cells. Preliminary studies with select cochaperone deletion mutants indicate that FHV RNA replication in S. cerevisiae does not depend on Hsp90 chaperone complex activity but is substantially reduced in the absence of Hsp70 chaperone activity, suggesting that FHV may use established cellular chaperone pathways to target and transport protein A to the appropriate intracellular membrane for viral RNA replication complex assembly (S. Weeks and D. Miller, unpublished data). The ability to selectively retarget functional FHV RNA replication complexes to alternative intracellular membranes (32) presents the opportunity to further investigate the role of both general and membrane-specific cellular chaperones in viral RNA replication complex assembly and function. ACKNOWLEDGMENTS We thank Gloria Wanty and Leslie Goo for assistance and David Markovitz, Alice Telesnitsky, and Yoichi Osawa for helpful comments on the research and manuscript. We thank Jorge Iniguez-Lluhi for providing plasmids and Paul Ahlquist for assistance with initial Drosophila S2 cell experiments. This work was supported in part by National Institutes of Health grant K08-AI01770. REFERENCES 1. Ahlquist, P., A. O. Noueiry, W. M. Lee, D. B. Kushner, and B. T. Dye. 2003. Host factors in positive-strand RNA virus genome replication. J. Virol. 77:8181–8186. 2. Arbeitman, M. N., and D. S. Hogness. 2000. Molecular chaperones activate the Drosophila ecdysone receptor, an RXR heterodimer. Cell 101:67–77. 3. Ball, L. A. 1995. Requirements for the self-directed replication of flock house virus RNA 1. J. Virol. 69:720–727. 4. Ball, L. A., and K. L. Johnson. 1998. Nodaviruses of insects, p. 225–267. In L. K. Miller and L. A. Ball (ed.), The insect viruses. Plenum Publishing Corporation, New York, N.Y. 5. Bijlmakers, M. J., and M. Marsh. 2000. Hsp90 is essential for the synthesis and subsequent membrane association, but not the maintenance, of the Src-kinase p56(lck). Mol. Biol. Cell 11:1585–1595. 6. Bukau, B., and A. L. Horwich. 1998. The Hsp70 and Hsp60 chaperone machines. Cell 92:351–366. 7. Bunch, T. A., Y. Grinblat, and L. S. Goldstein. 1988. Characterization and use of the Drosophila metallothionein promoter in cultured Drosophila melanogaster cells. Nucleic Acids Res. 16:1043–1061. 8. Caplan, A. J., D. M. Cyr, and M. G. Douglas. 1992. YDJ1p facilitates polypeptide translocation across different intracellular membranes by a conserved mechanism. Cell 71:1143–1155. 9. Caplen, N. J., J. Fleenor, A. Fire, and R. A. Morgan. 2000. dsRNA-mediated gene silencing in cultured Drosophila cells: a tissue culture model for the analysis of RNA interference. Gene 252:95–105. 10. Cartwright, P., T. Beilharz, P. Hansen, J. Garrett, and T. Lithgow. 1997. Mft52, an acid-bristle protein in the cytosol that delivers precursor proteins to yeast mitochondria. J. Biol. Chem. 272:5320–5325. 11. Clemens, J. C., C. A. Worby, N. Simonson-Leff, M. Muda, T. Maehama, B. A. Hemmings, and J. E. Dixon. 2000. Use of double-stranded RNA interference in Drosophila cell lines to dissect signal transduction pathways. Proc. Natl. Acad. Sci. USA 97:6499–6503.

J. VIROL. 12. Cutforth, T., and G. M. Rubin. 1994. Mutations in Hsp83 and cdc37 impair signaling by the sevenless receptor tyrosine kinase in Drosophila. Cell 77: 1027–1036. 13. Friesen, P. D., and R. R. Rueckert. 1981. Synthesis of black beetle virus proteins in cultured Drosophila cells: differential expression of RNAs 1 and 2. J. Virol. 37:876–886. 14. Funfschilling, U., and S. Rospert. 1999. Nascent polypeptide-associated complex stimulates protein import into yeast mitochondria. Mol. Biol. Cell 10:3289–3299. 15. Guarino, L. A., and P. Kaesberg. 1981. Isolation and characterization of an RNA-dependent RNA polymerase from black beetle virus-infected Drosophila melanogaster cells. J. Virol. 40:379–386. 16. Guinea, R., and L. Carrasco. 1990. Phospholipid biosynthesis and poliovirus genome replication, two coupled phenomena. EMBO J. 9:2011–2016. 17. Hachiya, N., T. Komiya, R. Alam, J. Iwahashi, M. Sakaguchi, T. Omura, and K. Mihara. 1994. MSF, a novel cytoplasmic chaperone which functions in precursor targeting to mitochondria. EMBO J. 13:5146–5154. 18. Hartl, F. U., and M. Hayer-Hartl. 2002. Molecular chaperones in the cytosol: from nascent chain to folded protein. Science 295:1852–1858. 19. Hoogenraad, N. J., L. A. Ward, and M. T. Ryan. 2002. Import and assembly of proteins into mitochondria of mammalian cells. Biochim. Biophys. Acta 1592:97–105. 20. Hu, J., and C. Seeger. 1996. Hsp90 is required for the activity of a hepatitis B virus reverse transcriptase. Proc. Natl. Acad. Sci. USA 93:1060–1064. 21. Hu, J., D. Toft, D. Anselmo, and X. Wang. 2002. In vitro reconstitution of functional hepadnavirus reverse transcriptase with cellular chaperone proteins. J. Virol. 76:269–279. 22. Hu, J., D. O. Toft, and C. Seeger. 1997. Hepadnavirus assembly and reverse transcription require a multi-component chaperone complex which is incorporated into nucleocapsids. EMBO J. 16:59–68. 23. Hung, J. J., C. S. Chung, and W. Chang. 2002. Molecular chaperone Hsp90 is important for vaccinia virus growth in cells. J. Virol. 76:1379–1390. 24. Johnson, K. L., and L. A. Ball. 1999. Induction and maintenance of autonomous flock house virus RNA1 replication. J. Virol. 73:7933–7942. 25. Lee, D. H., and A. L. Goldberg. 1998. Proteasome inhibitors: valuable new tools for cell biologists. Trends Cell Biol. 8:397–403. 26. Li, H., W. X. Li, and S. W. Ding. 2002. Induction and suppression of RNA silencing by an animal virus. Science 296:1319–1321. 27. Lindenbach, B. D., J. Y. Sgro, and P. Ahlquist. 2002. Long-distance base pairing in flock house virus RNA1 regulates subgenomic RNA3 synthesis and RNA2 replication. J. Virol. 76:3905–3919. 28. Loo, M. A., T. J. Jensen, L. Cui, Y. Hou, X. B. Chang, and J. R. Riordan. 1998. Perturbation of Hsp90 interaction with nascent CFTR prevents its maturation and accelerates its degradation by the proteasome. EMBO J. 17:6879–6887. 29. Marchler, G., and C. Wu. 2001. Modulation of Drosophila heat shock transcription factor activity by the molecular chaperone DROJ1. EMBO J. 20:499–509. 30. Miller, D. J., and P. Ahlquist. 2002. Flock house virus RNA polymerase is a transmembrane protein with amino-terminal sequences sufficient for mitochondrial localization and membrane insertion. J. Virol. 76:9856–9867. 31. Miller, D. J., M. D. Schwartz, and P. Ahlquist. 2001. Flock house virus RNA replicates on outer mitochondrial membranes in Drosophila cells. J. Virol. 75:11664–11676. 32. Miller, D. J., M. D. Schwartz, B. T. Dye, and P. Ahlquist. 2003. Engineered retargeting of viral RNA replication complexes to an alternative intracellular membrane. J. Virol. 77:12193–12202. 33. Momose, F., T. Naito, K. Yano, S. Sugimoto, Y. Morikawa, and K. Nagata. 2002. Identification of Hsp90 as a stimulatory host factor involved in influenza virus RNA synthesis. J. Biol. Chem. 277:45306–45314. 34. Ochel, H. J., K. Eichhorn, and G. Gademann. 2001. Geldanamycin: the prototype of a class of antitumor drugs targeting the heat shock protein 90 family of molecular chaperones. Cell Stress Chaperones 6:105–112. 35. Panaretou, B., C. Prodromou, S. M. Roe, R. O’Brien, J. E. Ladbury, P. W. Piper, and L. H. Pearl. 1998. ATP binding and hydrolysis are essential to the function of the Hsp90 molecular chaperone in vivo. EMBO J. 17:4829–4836. 36. Perez, L., R. Guinea, and L. Carrasco. 1991. Synthesis of Semliki Forest virus RNA requires continuous lipid synthesis. Virology 183:74–82. 37. Poels, J., A. Martinez, M. M. Suner, A. De Loof, S. J. Dunbar, and J. Vanden Broeck. 2004. Glucocorticoid-inducible gene expression vectors for use in Drosophila melanogaster Schneider 2 cells. Insect Mol. Biol. 13:205–211. 38. Pratt, W. B., and D. O. Toft. 2003. Regulation of signaling protein function and trafficking by the hsp90/hsp70-based chaperone machinery. Exp. Biol. Med. 228:111–133. 39. Price, B. D., M. Roeder, and P. Ahlquist. 2000. DNA-directed expression of functional flock house virus RNA1 derivatives in Saccharomyces cerevisiae, heterologous gene expression, and selective effects on subgenomic mRNA synthesis. J. Virol. 74:11724–11733. 40. Price, B. D., R. R. Rueckert, and P. Ahlquist. 1996. Complete replication of an animal virus and maintenance of expression vectors derived from it in Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 93:9465–9470. 41. Rapoport, T. A., B. Jungnickel, and U. Kutay. 1996. Protein transport across

VOL. 79, 2005

42.

43.

44. 45.

46. 47. 48.

the eukaryotic endoplasmic reticulum and bacterial inner membranes. Annu. Rev. Biochem. 65:271–303. Roe, S. M., C. Prodromou, R. O’Brien, J. E. Ladbury, P. W. Piper, and L. H. Pearl. 1999. Structural basis for inhibition of the Hsp90 molecular chaperone by the antitumor antibiotics radicicol and geldanamycin. J. Med. Chem. 42:260–266. Schmidt-Mende, J., E. Bieck, T. Hu ¨gle, F. Penin, C. M. Rice, H. E. Blum, and D. Moradpour. 2001. Determinants for membrane association of the hepatitis C virus RNA-dependent RNA polymerase. J. Biol. Chem. 276:44052– 44063. Schneemann, A., and D. Marshall. 1998. Specific encapsulation of nodavirus RNAs is mediated through the C terminus of capsid precursor protein alpha. J. Virol. 72:8738–8746. Schulte, T. W., S. Akinaga, T. Murakata, T. Agatsuma, S. Sugimoto, H. Nakano, Y. S. Lee, B. B. Simen, Y. Argon, S. Felts, D. O. Toft, L. M. Neckers, and S. V. Sharma. 1999. Interaction of radicicol with members of the heat shock protein 90 family of molecular chaperones. Mol. Endocrinol. 13:1435– 1448. Schulte, T. W., M. V. Blagosklonny, C. Ingui, and L. Neckers. 1995. Disruption of the Raf-1-Hsp90 molecular complex results in destabilization of Raf-1 and loss of Raf-1-Ras association. J. Biol. Chem. 270:24585–24588. Scotti, P. D., S. Dearing, and D. W. Mossop. 1983. Flock House virus: a nodavirus isolated from Costelytra zealandica (White) (Coleoptera: Scarabaeidae). Arch. Virol. 75:181–189. Sheffield, W. P., G. C. Shore, and S. K. Randall. 1990. Mitochondrial precursor protein. Effects of 70-kilodalton heat shock protein on polypeptide folding, aggregation, and import competence. J. Biol. Chem. 265:11069– 11076.

Hsp90 AND FHV RNA REPLICATION

6837

49. Sullivan, C. S., and J. M. Pipas. 2001. The virus-chaperone connection. Virology 287:1–8. 50. Supino-Rosin, L., A. Yoshimura, Y. Yarden, Z. Elazar, and D. Neumann. 2000. Intracellular retention and degradation of the epidermal growth factor receptor, two distinct processes mediated by benzoquinone ansamycins. J. Biol. Chem. 275:21850–21855. 51. Thulasiraman, V., and R. L. Matts. 1996. Effect of geldanamycin on the kinetics of chaperone-mediated renaturation of firefly luciferase in rabbit reticulocyte lysate. Biochemistry 35:13443–13450. 52. Towner, J. S., T. V. Ho, and B. L. Semler. 1996. Determinants of membrane association for poliovirus protein 3AB. J. Biol. Chem. 271:26810–26818. 53. Truscott, K. N., K. Brandner, and N. Pfanner. 2003. Mechanisms of protein import into mitochondria. Curr. Biol. 13:R326–R337. 54. Waxman, L., M. Whitney, B. A. Pollok, L. C. Kuo, and P. L. Darke. 2001. Host cell factor requirement for hepatitis C virus enzyme maturation. Proc. Natl. Acad. Sci. USA 98:13931–13935. 55. Whitesell, L., E. G. Mimnaugh, B. De Costa, C. E. Myers, and L. M. Neckers. 1994. Inhibition of heat shock protein HSP90-pp60v-src heteroprotein complex formation by benzoquinone ansamycins: essential role for stress proteins in oncogenic transformation. Proc. Natl. Acad. Sci. USA 91:8324–8328. 56. Wu, S. X., and P. Kaesberg. 1991. Synthesis of template-sense, single-strand Flockhouse virus RNA in a cell-free replication system. Virology 183:392– 396. 57. Young, J. C., N. J. Hoogenraad, and F. U. Hartl. 2003. Molecular chaperones Hsp90 and Hsp70 deliver preproteins to the mitochondrial import receptor Tom70. Cell 112:41–50.